Messy notes → Clear action items. Instantly. Paste any meeting notes, transcript, or text. Get summaries, action items with owners and deadlines. Auto-saved, searchable, with integrated to-do tracking. No bot. No subscription. No setup.
Security Analysis
high confidenceThe skill's requests and instructions are internally consistent with a meeting-notes + to‑do tracker: it requires no external credentials or installs and only instructs the agent to parse text and save files locally.
Name/description (extract summaries, action items, save/searchable notes, to-do tracking) match the SKILL.md and example files. No unrelated binaries, credentials, or external services are requested.
Instructions are strict about one-message output, file naming, and automatically saving an .md file and updating todo.md in the workspace. This is coherent for a note-taking/tracker skill, but it explicitly reads/writes workspace files (preferences, meeting-notes/, todo.md). That file I/O is within the stated purpose but is a privacy/side‑effect to be aware of—it will persist meeting contents unless you prevent or review it.
Instruction-only skill with no install spec, no external downloads, and no code files to execute. Lowest-risk install profile.
No environment variables, credentials, or config paths are required. The skill optionally checks for a preferences file in the workspace; this optional file-read is proportionate to the feature (customization).
always:false (normal). The skill expects to create and update files (meeting-notes/*.md and todo.md) in the agent workspace. That persistent storage is appropriate for a notes/todo tool but is a behavior users should explicitly accept because it stores potentially sensitive meeting content.
Guidance
This skill looks coherent and does what it says: parse pasted notes/transcripts, extract action items, and save a .md file plus a todo.md in your workspace. Before installing or using it, consider: (1) it will write persistent files (meeting-notes/* and todo.md) — review and control where your agent's workspace is located; (2) any sensitive meeting content you paste will be saved locally unless you delete it; (3) the skill will optionally read a preferences file from your workspace if present, which can alter behavior — inspect that file before relying on it; (4) there are no external network calls or credential requests declared, so no obvious exfiltration vectors in the SKILL.md. If you want extra caution, test with non-sensitive dummy notes first and confirm where files are written and who has access to the workspace.
Latest Release
v1.0.3
Version 1.0.3 - Major update to clarify and enforce the meeting extraction response format. - Added a "CRITICAL: RESPONSE FORMAT" section detailing that all meeting results, action items, filename, and add-to-to-do-list prompt must be presented in ONE single message. - Made numbered action items in chat and to-do prompt mandatory for all outputs. - Strict naming and file attachment requirements enforced (e.g., filenames must be date-first). - Listed explicit rules for what to always/never do to ensure clarity and reliability for users.
More by @jeffjhunter
Published by @jeffjhunter on ClawHub
